An empirical relationship between waste water content, density and biogas production in reconstituted municipal solid waste: a laboratory scale experiment S. GHOLAMIFARD*, R. EYMARD** * Veolia Environnement, Environmental Services Research Center, Zone Portuaire de Limay, 291 Avenue Dreyfous Ducas, 78520, Limay, France, ** LETEM, Marne-La-Vallée University, Cité Descartes 5, bd Descartes, Champs sur Marne, 77454 Marne La Vallée, Cedex 2, France

SUMMARY: This paper describes the laboratory tests carried out to study the influence of water content and density of waste, as two important factors which influence biodegradation (Powrie, 1999 ; Zornberg, 1999 ; Lanini, 1998 ; Ademe, 2001), on biogas production and development of microorganisms in waste. Four levels of density and three levels of saturation are studied in 13 laboratory scale cells of about 2 liters. The four levels of density are concerned to 2, 4, 8 and 13m of wastes in a landfill (Olivier, 2003) and the three levels of saturation are: minimum saturation (Schulze, 1961; Haug, 1993; Palmisano and Barlaz, 1996; Buivid et al., 1981; Noble et al., 1991; Gurijala and Suflita, 1993), field capacity and maximum saturation. The experimental cells are kept at constant temperature of about 33-35°C. In this paper we present first the preparation of experimental cells; waste compositions, the protocol of compaction and leachate injection. We present then the results of biogas production and development of microorganisms in the cells as a function of water content and density and we discuss the results.

1. INTRODUCTION The main objectives of bioreactor landfills are to accelerate anaerobic degradation of waste in order to minimize the environmental impacts, to optimize biogas production and to minimize the time of waste stabilization as well as the costs and time of monitoring of landfill sites after operation. These objectives could not be achieved without enough knowledge of hydraulic, thermal and biological parameters and processes of bioreactor landfills and the effects of each of them on the others. Site observations and laboratory experiments as well as numerical models could help to develop the knowledge of these phenomena and processes. In our last paper (Gholamifard et al., 2008) we studied the coupled hydro-thermo-biological behavior of bioreactor landfills with and without leachate recirculation in anaerobic phase. We presented a numerical coupled model based on microbial activity, as a function of waste saturation and temperature. The laboratory experiments help us to know the effect of important parameters and processes as saturation, density of waste and leachate recirculation on anaerobic degradation, biogas production and development of microorganisms in waste. These works are carried out in parallel, during my PhD thesis in Cemagref d’Antony, and in a complementary manner to improve our knowledge of biodegradation processes.

In this paper we present the materials and experimental methods, the results of biogas production and development of microorganisms in the cells as a function of water content and density and finally we discuss the results.

2. MATERIALS AND METHODES We studied the effects of saturation and density, and so indirectly the effect of porosity, of waste on degradation of solid waste and biogas production in 13 PVC test cells filled with wastes, considered as bioreactors (Figure 1). About 25kg of solid waste have been shredded into 1-2cm pieces (Barlaz et al. 1992), using the waste proportions in France, proposed by Ademe (1993). The leachate used to saturate the wastes comes from a landfill site in France in stable methanogenic phase, which is kept at a temperature of 4°C before injection into experimental cells. The wastes are placed and then compacted in test cells in four levels of dry density (Table 1) related to four levels of Δh in Figure 1, and are saturated to three levels: 35% of field capacity (Smin), field capacity (FC) and maximum saturation (Smax). The levels of dry density are related to wastes at 2, 4, 8 and 13m depth (Olivier 2003). The nomination of the cells with related saturation and density are presented in Table 2.

Biogas bag

Temperature sensor

Biogas vent

Waste Leachate tap

Figure (1) Experimental cell and dimensions

Table 1. Compaction levels and related densities Density (kg/m3) 450 490 540 580

Δh (cm) 0.5 4.5 5.5 6.5

Table 2. Nomination of the cells with related saturation and density Maximum saturation Field capacity 35% of field capacity Cell-1: (Smax, 450 kg/m3)

Cell-5: (FC, 450 kg/m3)

Cell-9: (Smin, 450 kg/m3)

Cell-2: (Smax, 490 kg/m3)

Cell-6: (FC, 490 kg/m3)

Cell-10: (Smin, 490 kg/m3)

Cell-3: (Smax, 540 kg/m3)

Cell-7: (FC, 540 kg/m3)

Cell-11: (Smin, 540 kg/m3)

Cell-4: (Smax, 580 kg/m3)

Cell-8: (FC, 580 kg/m3)

Cell-12: (Smin, 580 kg/m3)

Dry cell: Cell-13 (Dry, 490 kg/m3)

2.1 Protocol In order to accelerate the methanogenic phase of degradation in our cells, we have added some methanogenic sludge to the wastes before putting them in test cells. We accelerated this phase adding 180gr of humid sludge with 87.7% of water content to our wastes. We added 100ml of leachate to sludge in order to liquefy it and we added it to the wastes manually trying to homogenize the sludge in wastes. To achieve the saturation levels for minimum saturation, field capacity and maximum saturation, we followed these steps: 1) add the sludge to wastes and try to homogenize it in waste, 2) fill the cell in three steps applying the vertical constraint to achieve each level of density, 3) inject leachate from the bottom tap until the saturation of the whole waste and measure the injected volume. Leave then the bottom tap open during about half an hour and measure the exiting leachate, 4) put the PVC cap, apply the charge up to achieve desired density ( Δh ), fix the cap level and measure the exiting leachate, For the cells with maximum saturation 5) inject the same leachate exited from the test cell up to the total saturation and measure the injected leachate, For the cells with field capacity 5) inject the same leachate exited from the test cell up to the total saturation and leave them as much time as necessary until there is no more leachate exiting from the tap (according to the definition of field capacity), measure of all the injected and exited leachates, For the cells with minimum saturation 5) add 35% of the volume of leachate injected in the cells with field capacity in the cells with a minimum saturation, before filling them and compacting the waste. The last step for all cells is that we closed the leachate tap, attached the gas bag, weighted the test cells and put them at constant temperature of about 33-35°C. The water content is calculated using the following equation: Water − Content =

Mh − Md Mh

where Md is dry mass and Mh is the humid mass of waste after leachate injection.

(1)

2.2 Analytical methodes Biogas produced in test cells were analyzed by gas chromatography (μGC CP2003P Varian) equipped with thermal conductivity detectors (TCD). Helium was used as carrier gas. The volume of the biogas was measured using Archimedes' law. To find this weight, we pushed the gas bag in a basin of water placed on a balance. The weight of water displaced is the volume of gas. We subtracted of this weight the weight of the empty bag.

3. RESULTS AND DISCUSSION 3.1 Total biogas production The results of total biogas production after one year are presented in Figure 2. The surface of the circles shows the volume of biogas produced. We represent the results as functions of water content and solid density. The solid density is chosen because the wastes are compacted at first at their solid state, with some constitutional water content. The first remark is that the more humid and less dense waste produces much more biogas than the dryer and denser waste. As we can see in Figure 2, the cells which produced the most biogas are cells 5 (95L of biogas), 1 (60L of biogas), 8 (48L of biogas) and 2 (40L of biogas). Cells 4 and 6 have produced 15L and 16L of biogas respectively which is more than the other cells. The dry cells have produced almost no biogas. In these cells the biogas production was stopped about one or two months after their establishment. We can see in these results that water content is a key factor of biogas production and degradation in these cells. Regardless to the density, the cells with less than 40% of water content do not produce biogas. However as we can see in Figure 2, even in dryer cells the effect of density on biogas production is significant. The effects of waste density and water content are shown in Figure 2 by dashed lines. The results of biogas production in different cells show that these values are nearer to in situ values than the potential of biogas production from waste. Considering that the potential values obtain in ideal conditions, it shows that our test cells are representative of in situ conditions where the variations of density and water content influence biogas production.

Water content (%)

Density effect

Water content effect

Cell-5 : 95L of biogas Cell-8 : 48L of biogas Cell-12 : 4L of biogas

Solid density (kg/m3) Figure (2) Biogas production in test cells after one year

A less biogas production in denser waste could be a result of a less contact surface between solid substrates and water, as well as smaller pores which slow down the liquid flow and increase the inhibition possibility by VFA accumulation. A more possible preferential flow during leachate injection in denser cells could be another reason for these results which leads to a less homogenized saturation in denser waste. 3.2 Cumulated biogas production and biogas composition The results of cumulated biogas production over time are presented in Figure 3 in all test cells. As we can see in these figures, the biogas production is stopped in all cells after a few months except four cells: 1, 2, 5 and 8. Two kinds of behaviors in these four cells could be distinguished: the behavior of cells 1 and 5 are very close to each other and biogas production starts very rapidly in these cells after a few times, which could represent a temporary inhibition of methanogenic phase. The behavior of Test cells 2 and 8 shows that biogas production continues vary regularly in these cells in one year. Among the cells where biogas production stopped after a few months, cells 4 and 6 produced more biogas during a longer time. Figure 4 shows the composition of biogas (CO2 and CH4) in these four cells. We could recognize different phases of degradation according to the biogas composition (Farquhar and Rovers, 1973), and the evolution of biogas during one year in different cells. The four phases are: phase I, aerobic phase in the presence of oxygen, in this phase biogas is composed mainly of N2 (80%) and O2 (20%). Phase II, anaerobic acid phase, where N2 decreases and CO2 and H2 increase to achieve 80% and 20% of biogas, respectively. Phase III, anaerobic accelerated methanogenic phase where CO2 and H2 decrease to values less than 50% for CO2, methane production increases in this phase and there is more than 50% of methane in biogas. Phase IV is anaerobic decelerated methanogenic phase where the methane and carbon dioxide compositions in biogas remain approximately constant more and less than 50%, respectively. As we can see in Figure 4, cell 5 reaches the methanogenic phase faster than the other cells. Methanogenic phase lasts more in cell 1 than in the other cells which explains the sudden increase of cumulated biogas production which is observed in Figure 3. The dashed lines in cell 1 show a decelerated methanogenic phase which could be due to local VFA accumulation in this test cell. Biogas production in cells 4 and 6 stoped after 2-3 months in the methanogenic phase. Cumulated biogas production (L) 100 Cell-5

Biogas (L)

80 Cell-1 60 Cell-8 Cell-2

40

20

0 0

50

100

150

200

250

300

350

time (d)

Figure (3) Cumulated biogas production in test cells after one year

Cell-1

80

CH4

Inhibition?

60 40 20

CO2

0

80 CH4 60 40 CO2

20 0

0

50

100

150 200 Time (d)

250

300

350

0

Biogas composition (%)

CH4

40 CO2

20

100

150 200 Time (d)

250

300

350

300

350

100

80 60

50

Cell-8

Cell-2

100 Biogas composition (%)

Cell-5

100 Biogas composition (%)

Biogas composition (%)

100

80 CH4 60 40 CO2 20 0

0 0

50

100

150 200 Time (d)

250

300

350

0

50

100

150 200 Time (d)

250

Figure (4) Biogas composition in test cells after one year

3.3 Leachate injection and definition of an optimum zone of biogas production To study the effect of leachate injection in waste, we chose a dry test cell, cell 10, which did not produce any biogas for a long time. We have injected leachate two times in cell 10. Two weeks after the second injection, the biogas production started in this cell. This could be because of the dilution effect of the second injection which decreased the high VFA concentration in cell 10 produced after the first injection, or just a lag time for biogas production. We calculated the new water content of this test cell, which we called cell 10* after leachate injection, and we place it on the same figure as Figure 2. The results are shown in Figure 5 for all test cells, where cell 10* is shown in pink. As we can see in this figure, the water content of cell 10 has increased from 35% to 50% after two periods of leachate injection. These results let us define an empirical relationship between water content, density and biogas production in reconstituted waste. It seems that the blue zone in Figure 5 is an optimum zone of biogas production regarding to water content and density of wastes, in our experimental conditions.

Water content (%)

Cell-5 : 95L of biogas Cell-8 : 48L of biogas Injection Cell-12 : 4L of biogas Solid density (kg/m3) Figure (5) Biogas production after one year in all test cells

To provide the possibility of a better interpretation of results we have taken out biological analysis in leachate samples from test cells, using the FISH (Fluorescence In Situ Hybridization) technique. The results of microbiological analyses show that cell 5 has a leachate sample which is the most concentrated in methanogenic biomass. This is the test cell that produced the most biogas. Leachate of test cells 1, 2 and 8 are also very concentrated in methanogenic biomass and fermentative bacteria. All these test cells produced lots of biogas and as we can see in Figure 5 they are the less dense (except cell 8) and the most humid test cells. In the leachate of test cell 12, although it was 5 times centrifuged, almost no methanogenic biomass was observed. This is one of the densest and driest test cells. The results of microbiological analyses are presented in Figure 6, for cells 5 and 12. We can observe the methanogenic populations in red and the fermentative bacteries in green.

Cell 12

Cell 5

5µm

5µm

Figure (6) The results of microbiological analyses of test cells 5 and 12

5. CONCLUSIONS We can summarize the conclusions of this experimental work as follow: • We found a relationship between water content of waste, density and biogas production and microbial evolution. We observed that as the waste is denser and dryer, they produce less biogas. The microbial evolution and activity are also less in dryer and denser waste. • The water content of waste is a key factor of biogas production and degradation. There is a very low biogas production in waste with water content under 40% (water volume /waste volume), independently to density. The density of waste is also a very important factor of degradation and biogas production. It seems that high density lessens biogas production. Density effect is observed even in dry wastes. The density effect could be because of VFA accumulation, as a result of less liquid flow in pores in higher densities. And may be in higher density, the surface of biodegradable solids in contact with liquid is smaller. A more possible preferential flow during leachate injection in denser cells could be another reason for these results which leads to a less homogenized saturation in denser waste. As in the literature there are many contradictory results about the effect of density on biodegradation, these results should be verified in future works. • The results of biogas production in different cells show that our test cells are representative of in situ conditions where the variations of density and water content influence biogas production. • There is an increasing relationship between water content, biogas production and microbial evolution and a decreasing relationship with density, for the waste with water content higher than 45%. • The maximal values of biogas production are related to water content between 45 and 60% and density between 400 and 550 kg/m3 in our test. It was technically impossible to achieve water content higher that 60% in our test cells because gas pressure pushes leachate out through the biogas vent. • It is very important to saturate and inoculate the waste homogeneously. The non degraded zones observed in test cells were probably improperly saturated and/or inoculated by methanogenic sludge. • Channelling flow is observed during leachate injection, especially when discharge rate is high, and prevents a homogeneous saturation of wastes. Injection discharge should be well controlled to prevent channelling flow. • Non degraded zones are mostly observed at the top of the test cells; however the waste situated at the bottom is usually degraded. This could be because of the vertical leachate flow which forms a more humid zone at the bottom of the cells. A non degraded zone at the bottom could be because of the VFA accumulation and hydrolysis inhibition of the humid zones, or because of a heterogeneous saturation and inoculation of waste in these zones. • Leachate injection influences more the dryer waste than the actually humid ones; it activates biogas production in dryer wastes very fast. This could be because in humid waste the rapidly biodegradable components are already degraded.

REFERENCES Ademe (1993). MODECOM™. Méthode de Caractérisation des Ordures Ménagères. Connaître pour agir Manuel de base, 64 pages + complément. Ademe, B. (2001). Guide pour le dimensionnement et la mise en oeuvre des couvertures de sites de stockage de déchets ménagers et assimilés. A. Editions. Paris: 157 pages + annexes. Barlaz M. A., Robert K . Hamt and Daniel M . (1992). Schaefer, Microbial, Chemical and Methane Production Characteristics of Anaerobically Decomposed Refuse With and Without Leachate Recycling, Waste Management and Research 10, 257-267 Farquhar G.J., Rovers F.A., (1973). Gas Production During Refuse Decomposition. Water, Air and Soil Pollution 2, pp 483-495. Gholamifard S., Eymard R. and Duquennoi C. (2008). Modeling Anaerobic Bioreactor Landfills in Methanogenic Phase: Long Term and Short Term Behaviors, Water Research, DOI: 10.1016/j. waters. Lanini S. (1998). Analyse et Modélisation des Transferts de Masse et de Chaleur au Sein des Décharges d’Ordures Ménagères. Institut National Polytechnique des Toulouse, PhD Thesis, Toulouse, 148 p. Noble, J.J. and Arnold, A.E., (1991). Experimental and Mathematical Modeling of Moisture Transport in Landfills, Chemical Eng. Comm., 100:95-111. Olivier F., (2003). Settelement of municipal solid waste in landfill: Site monitoring to modelling, PhD thesis, Lirigm laboratory, Université Joseph Fourier - Grenoble I, Gronoble, France. Powrie W., Beaven, R.P. et Harkness, R.M. (1999). Applicability of soil mechanics principles to household waste. Proc. Sardinia 99, 7th International Landfill Symposium, Cagliari, Vol. III, pp. 429-436. Zornberg, J.G., Jernigan, B.L., Sanglerat, T.H. et Cooley, B.H. (1999). Retention of free liquid in landfill undergoing vertical expansion. Journal of Geotechnical and GeoEnvironmental Engineering, Vol. 125, n° 7, pp. 583-594.